Intelligent Analysis Method of Indicator Diagram Based on Convolutional Neural Network CNN
A convolutional neural network-based intelligent analysis and monitoring model for marine diesel engine indicator diagrams is constructed to address the issue of uncertainty in the current analysis and evaluation of indicator diagrams,which mostly rely on manual work and rely on the business capabilities and experience of engine management personnel.The model includes 5 layers of convolutional neural networks and 2 layers of fully connected neural networks,and adopts various optimization algorithms and hardware acceleration using GPU.Through training and experiments,the results show that the training accuracy of the indicator diagram is 99.5%,the recognition accuracy is 95.9%,and the average recognition time is 0.032 seconds,which verifies the reliability and accuracy of the model.It can achieve intelligent analysis of the indicator diagram,meet the accuracy requirements of intelligent monitoring of diesel engine combustion conditions,and provide support for the intelligent management of ship power systems
NETL
NSTL
万方数据
